Journal of Controlled Release 92 (2003) 173187

www.elsevier.com / locate / jconrel

Polymer degradation and in vitro release of a model protein

from poly( D,L-lactide-co-glycolide) nano- and microparticles
Jayanth Panyam a , Manisha M. Dali b ,1 , Sanjeeb K. Sahoo a , Wenxue Ma a ,
Sudhir S. Chakravarthi a , Gordon L. Amidon b , Robert J. Levy c ,
Vinod Labhasetwar a,d , *
a

Department of Pharmaceutical Sciences, 986025 University of Nebraska Medical Center, Omaha, NE 68198 -6025, USA
b
College of Pharmacy, University of Michigan, Ann Arbor, MI 48109, USA
c
Division of Cardiology, Children s Hospital of Philadelphia, Department of Pediatrics,
University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA
d
Department of Biochemistry and Molecular Biology, University of Nebraska Medical Center, Omaha, NE 68198, USA
Received 5 June 2003; accepted 27 June 2003

Abstract
The objective of the study was to investigate the effect of particle size of nano- and microparticles formulated from
poly( D,L-lactide-co-glycolide) (50:50 PLGA) on polymer degradation and protein release. Since the surface area to volume
ratio is inversely proportional to the particle size, it is hypothesized that the particle size would influence the polymer
degradation as well as the release of the encapsulated protein. PLGA nano- and microparticles of approximate mean
diameters of 0.1, 1 and 10 mm, containing bovine serum albumin as a model protein, were formulated using a multiple
water-in-oil-in-water emulsion solvent evaporation technique. These particles were incubated at 37 8C in phosphate-buffered
saline (pH 7.4, 154 mM) and the particles were characterized at various time points for molecular weight of polymer,
surface-associated polyvinyl alcohol content (PVA), and the particle surface topology using scanning electron microscopy.
The supernatants from the above study were analyzed for the released protein and PVA content. Polymer degradation was
found to be biphasic in both nano- and microparticles, with an initial rapid degradation for 2030 days followed by a slower
degradation phase. The 0.1 mm diameter nanoparticles demonstrated relatively higher polymer degradation rate (P,0.05)
during the initial phase as compared to the larger size microparticles (first order degradation rate constants of 0.028 day 21 ,
0.011 day 21 and 0.018 day 21 for 0.1, 1 and 10 mm particles, respectively), however the degradation rates were almost
similar (0.008 to 0.009 day 21 ) for all size particles during the later phase. All size particles maintained their structural
integrity during the initial degradation phase; however, this was followed by pore formation, deformation and fusion of
particles during the slow degradation phase. Protein release from 0.1 and 1 mm particles was greater than that from 10 mm
size particles. In conclusion, the polymer degradation rates in vitro were not substantially different for different size particles
despite a 10- and 100-fold greater surface area to volume ratio for 0.1 mm size nanoparticles as compared to 1 and 10 mm
size microparticles, respectively. Relatively higher amounts of the surface-associated PVA found in the smaller-size

*Corresponding author. Tel.: 11-402-559-9021; fax: 11-402-559-9543.

E-mail address: vlabhase@unmc.edu (V. Labhasetwar).
1
Present address: Pharmaceutical Research Institute, Bristol-Myers Squibb, New Brunswick, NJ 08903, USA.
0168-3659 / 03 / $ see front matter 2003 Elsevier B.V. All rights reserved.
doi:10.1016 / S0168-3659(03)00328-6

174

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

nanoparticles (0.1 mm) as compared to the larger-size microparticles could explain some of the observed degradation results
with different size particles.
2003 Elsevier B.V. All rights reserved.
Keywords: Sustained release; Biodegradable polymers; Drug delivery; Particle size; Polyvinyl alcohol

1. Introduction
Poly ( D,L-lactide-co-glycolide) (PLGA) nano- and
microparticles are being extensively investigated for
sustained delivery of therapeutic agents including
DNA, proteins and low molecular weight therapeutic
agents [1,2]. For a number of these applications,
especially those involving potent drugs and macromolecules, it is important to be able to control the
rate and the duration of release of the entrapped
therapeutic agent. Previous studies by others have
shown that by modifying different formulation parameters, the release of the entrapped therapeutic
agent from nano- and microparticles can be altered
[39]. These formulation parameters include polymer and protein molecular weight, composition,
formulation method, particle size and the type of
emulsifier used. Release of the entrapped therapeutic
agent from PLGA matrix has been found to occur
through diffusion-cum-degradation-mediated processes [5]. It has been demonstrated that during the
early phases, the release of the entrapped therapeutic
agent occurs mainly through its diffusion in the
polymer matrix while during the later phases, the
release is mediated through both diffusion of the
therapeutic agent and the degradation of the polymer
matrix itself. Thus, degradation rate of the polymer
matrix is an important determinant of the in vitro
release of the therapeutic agent from PLGA matrices
[10].
Various factors including the device morphology,
polymer composition, tacticity, hydrophobicity / hydrophilicity, molecular weight and presence of other
additives have been shown to affect the degradation
rate of PLGA matrices [1013]. PLGA degradation
occurs through a process of autocatalytic hydrolysis
of the ester bonds [10]. The acidic (lactic acid and
glycolic acid) monomers and oligomers formed
catalyze the further degradation of the parent polymer. Thus, any factor that influences the formation
and / or retention of the acidic monomers in the
particles could affect the polymer degradation rate

and the in vitro release of the entrapped therapeutic

agent.
Particle size is an important parameter that could
affect the degradation of the polymer matrix [12]. As
the particle size is reduced, surface area to volume
ratio increases, resulting in a large surface area
available for the buffer penetration into the particles
and also for the faster escape of the polymer
degradation products from the smaller sizednanoparticles. Also, larger surface area of nanoparticles should result in higher in vitro release of the
encapsulated therapeutic agent. However, the amount
of surface-associated residual poly(vinyl alcohol)
(PVAused as an emulsion stabilizer during formulation of nano- and microparticles) could be higher
with smaller-size particles [14], and this in turn
could influence the rate and extent of protein release
from different size particles [15]. Thus, it was
hypothesized that there may be a more complex
relationship between particle size, polymer degradation and in vitro protein release. The objective of
the present study was therefore to determine the
effect of particle size on polymer degradation and
protein release characteristics of nano- and microparticles.
2. Materials and methods

2.1. Materials
Bovine serum albumin (BSA, Fraction V) and
PVA (average molecular weight 30,00070,000)
were purchased from Sigma (St. Louis, MO, USA).
PLGA (50:50 lactideglycolide ratio, 120,000 Da)
was purchased from Birmingham Polymers (Birmingham, AL, USA). All salts and buffers were of
reagent grade. Organic solvents were of HPLC
grade.

2.2. Formulation of nano- and microparticles

Nano- and microparticles containing BSA as a

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

model macromolecule were formulated using a double emulsionsolvent evaporation technique as described previously [16]. Specific formulation parameters and conditions for the preparation of different
size particles are listed in Table 1. An aqueous
solution of BSA (50 mg / 0.5 ml) was emulsified into
PLGA solution in methylene chloride (see Table 1
for volumes and concentrations used) using a probe
sonicator (55 W for 2 min) (Sonicator XL, Misonix, Farmingdale, NY, USA). The water-in-oil
emulsion thus formed was further emulsified into an
aqueous solution of PVA (volumes and concentrations are mentioned in Table 1) by sonication,
vortexing or using a tissue homogenizer (Tissue
Tearor , Fisher Scientific, Pittsburgh, PA, USA) for
5 min to form a multiple water-in-oil-in-water emulsion. The conditions for emulsification and the
formulation compositions were optimized to obtain
nano- or microparticles of desired diameters as per
our previously published protocols [17]. The multiple emulsion was stirred for |18 h at room temperature followed by 1 h in a desiccator under vacuum to
evaporate any residual methylene chloride.
Nanoparticles were recovered by ultracentrifugation
at 25,000 rev. / min (Optima LE-80K, Beckman,
Palo Alta, CA, USA) whereas microparticles were
recovered by normal centrifugation at 3200 rev. / min
(Sorvall RTH-750, DuPont, Newton, CT, USA). The
particles were washed two times to remove PVA and
unentrapped BSA and then lyophilized (Sentry,
Virtis, Gardiner, NY, USA) for 48 h to obtain a dry
powder.

2.3. BSA loading in nano- and microparticles

Protein loading in nano- and microparticles was
determined by analyzing the washings from each
formulation for the protein that is not encapsulated.
Protein content was measured by BCA protein assay
kit (Pierce, Rockford, IL, USA). The amount of
protein encapsulated in nano- and microparticles was
determined from the total amount of protein added
and the amount of protein that was not encapsulated.

175

ments Division, Santa Barbara, CA, USA). Optical

microscopy (Olympus, Lake Success, NY, USA) was
used to measure the size distribution and diameter of
other size particles.

2.5. Polymer degradation and protein release

About 4 mg of nano- and microparticles were
incubated with 1 ml of phosphate-buffered saline
(pH 7.4, 154 mM) in a 1.5-ml Eppendorf tube at
37 8C in a water bath rotary shaker at 100 rev. / min
(Model 2564, Forma Scientific, Marietta, OH, USA).
Three replicate tubes were used for each time point,
and the tubes were sampled at various time points.
The samples were centrifuged at 14,000 rev. / min
using a microcentrifuge (Eppendorf 5417R, Brinkmann Instruments, Westbury, NY, USA) for 15 min
and the pellet from each sample was lyophilized and
then used to determine the molecular weight of the
polymer and the residual PVA associated with nanoand microparticles. The supernatant from each of the
above samples was analyzed for the amount of BSA
released and the PVA content. Protein levels were
measured using the BCA protein assay kit. The pellet
was also analyzed for the shape and surface morphology of degraded nano- and microparticles by
scanning electron microscopy (SEM).

2.6. Effect of particle size on stability of protein

The protein released from different sized nanoand microparticles was analyzed using SDSPAGE.
About 5 ml of the released protein samples were
mixed with 25 ml of Laemellis buffer (BioRad,
Hercules, CA, USA) and boiled for 5 min. The
samples were then subjected to electrophoresis at 40
mA current in a vertical slab Ready Gel (12%,
BioRad). Proteins were visualized using the silver
staining kit (BioRad). Density of the protein bands
on the gel was analyzed using GelExpert software
(Nucleotech, San Mateo, CA, USA). The areas of the
bands were kept constant for the different gels and
the intensity of the bands were normalized to the
background of each gel.

2.4. Particle size distribution

2.7. Molecular weight determination
Particles of 0.1 mm diameter were characterized
for size distribution using a dynamic laser defractometer (NICOMP, Model 370, Hiac / Royco Instru-

The pellet obtained from the centrifugation step

was incubated in 1 ml of chloroform for 24 h and

176

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

then centrifuged at 14,000 rev. / min (Eppendorf

microcentrifuge) to remove the undissolved fraction
(BSA and PVA). The supernatant was then filtered
through a 0.1 mm filter (0.2 mm, Acrodisc , Aldrich,
Milwaukee, WI, USA) and then used for molecular
weight determination. A gel permeation chromatography system coupled to a laser light scattering
detector (DAWN DSP , Wyatt Technology, Santa
Barbara, CA, USA) and to interferometric refractometer (Optilab DSP, Wyatt Technology) was used
to determine the weight average molecular weight of
the polymer samples. A 0.5-ml sample of each was
7.83300 mm)
injected into an Ultrastyragel (10 3 A,
(Waters, Milford, MA, USA) column and the peaks
were resolved using chloroform as a mobile phase at
a flow rate of 1 ml / min. Initial experiments were
performed to analyze the specific refractive index
(dn / dc) of the PLGA polymer used for the formulation of nano- and microparticles according to the
manufacturers instructions (Wyatt Technology).

2.8. Analysis of PVA content

The residual PVA associated with nano- and
microparticles was determined by a previously reported method [15]. In brief, 2 mg of the lyophilized
particle sample was treated with 2 ml of 0.5 N
NaOH for 15 min at 60 8C. Samples were neutralized
with 900 ml of 1 N HCl and the volume of each
sample was adjusted to 5 ml with distilled water. To
each sample, 3 ml of 0.65 M solution of boric acid,
0.5 ml of I 2 / KI (0.05 M / 0.15 M) and 1.5 ml of
distilled water were added and the absorbance of the
samples were measured at 690 nm after 15 min
incubation. A standard plot of PVA was prepared
under identical conditions. The supernatants from the
released samples were also assayed using the above
procedure. To ensure that the oligomers (lactic and
glycolic acids) formed due to the degradation of
nano- and microparticles do not interfere in the PVA
assay, a standard plot for PVA was carried out in the
presence of 1 mg / ml of each of the oligomers. The
concentration of oligomers used in the assay was
based on the amount that would be formed with the
complete degradation of nano- or microparticles. The
results demonstrated that there was no interference of
oligomers in the PVA assay.

2.9. Scanning electron microscopy ( SEM)

characterization of nano- and microparticles
SEM was used to determine the surface topography and the morphology of the degrading particles.
A lyophilized sample of the particles obtained at
different time points was placed on a double stick
tape over aluminum stubs to get a uniform layer of
particles. Samples were gold-coated using a sputter
gold coater (Denton Vacuum, Cherry Hill, NJ, USA)
at 40 mA current and 50 mTorr pressure for 200 s at
Gold-coated particle samples
a thickness of 300 A.
were cooled over dry ice prior to SEM observations
to avoid their melting under high magnification due
to the electron beam exposure (Hitachi S570, Cherry
Hill, NJ, USA).

2.10. Statistical analysis

Results were expressed as mean6S.E.M. The
linearity of a first order fit for polymer degradation
was determined by calculating the root mean square
value (RSQ) for polymer molecular weight versus
time plot. Plots with RSQ value |0.9 were considered as linear. The differences in the polymer
degradation rate constants for different size particles
were analyzed by ANOVA. The significance of
differences in the individual rate constants were
analyzed by Tukeys pairwise comparisons.

3. Results and discussion

3.1. Formulation and characterization of nanoand microparticles

Different formulation parameters used for the
preparation of nano- and microparticles and their
characterization for mean particle size, protein loading and residual PVA are described in Tables 1 and
2. For 0.1 mm nanoparticles, more than 80% of
particles were in the size range of 80160 nm. For 1
mm microparticles, 80% of the particles were in the
size range of 8001200 nm and for 10 mm microparticles, more than 85% of the particles were in
the size range of 812 mm.

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

177

Table 1
Formulation of different size nano- and microparticles
Particle size

PLGA solution

PVA solution

Energy source

Actual diameter

Particle size average

0.1 mm
1 mm
10 mm

5.0 ml, 3.0% w / v

2.5 ml, 6.0% w / v
5.0 ml, 3.0% w / v

40 ml, 2.5% w / v
35 ml, 0.5% w / v
30 ml, 2.5% w / v

Sonicator
Homogenizer
Vortex

0.1165.0 mm
1.160.1 mm
9.460.2 mm

z
Number
Number

3.2. Polymer degradation in nano- and

microparticles
Particle size of the nano- and microparticles is an
important formulation property that influences other
pharmaceutical and cellular uptake properties of the
particulate delivery system. We have previously
shown that 0.1-mm-sized nanoparticles have about
10-fold higher intracellular uptake in an in vitro cell
culture model compared to 10-mm-sized microparticles [16]. Similarly, we have shown that the lower
sized nanoparticles have higher tissue uptake compared to microparticles [17]. The size of nanoparticles also influences their ability to act as drug / gene
delivery vehicles. Recently, we have shown that
,0.1 mm nanoparticles have about 27-fold higher
transfection compared to .0.1 mm size nanoparticles, which were fractionated from a nanoparticle
formulation by a filtration technique [18]. Decreasing
the particle size leads to a large increase in surface
area per unit volume of the particles. The surface
area to volume ratio of a microparticle is given by
6 /d where d is the diameter of the particle (10 mm)
and is equal to 0.6 mm 21 . For a 0.1 mm nanoparticle,
this ratio is equal to 60 mm 21 . Thus by reducing the
particle size from 10 mm to 0.1 mm, there is a
100-fold increase in surface area for a given weight
of the particles. This higher surface area associated
with the nanoparticle should theoretically facilitate
higher buffer contact / penetration into particles and
at the same time allow faster diffusion of monomers
and oligomers formed as a result of polymer degraTable 2
Characterization of different size nano- and microparticles
Particle
size

BSA loading
(% w / w)

Protein loading
efficiency (%)

Residual PVA
(mg / mg)

0.1 mm
1 mm
10 mm

22.8
22.6
21.9

90
91
88

51.460.2
24.061.6
33.861.1

dation out of the particles. Since PLGA degradation

has been shown to be catalyzed by acidic monomers
and oligomers, we hypothesized that there could be a
difference in the degradation rates of different sized
particles.
Initially, the specific refractive index (dn / dc)
value for the polymer in chloroform was determined
as per the method suggested by the manufacturer
(Wyatt Laboratories) and was found to be 0.033
ml / g. This was then used to determine the molecular
weight of the polymer samples from the degradation
study by laser light scattering. The polymer degradation profiles were found to be similar for particles
of all sizes (Fig. 1). The polymer degradation was
biphasic with an initial rapid degradation over the
first 2030 days followed by a slower rate of
degradation over the next 6070 days. The polymer
degradation rate constants were calculated from the

Fig. 1. The polymer molecular weight loss of degrading PLGA

nano- and microparticles. Nano- and microparticles were incubated in PBS and the polymer molecular weight was analyzed by
gel permeation chromatography coupled to a light scattering
detector. Each data point is a mean of three values. Standard error
was ,10% of the mean in all cases. Insert is a log (molecular
weight) vs. time plot of the above data.

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

178

formula: rate constant5 22.3033slope of log molecular weight versus time plot. The polymer degradation rate was faster for the 0.1 mm nanoparticles
compared to the microparticles during the initial
rapid degradation phase (P,0.05). However, there
was no significant difference between the degradation rates of 1 mm and 10 mm particles. The rate
constants for the first rapid decline phase were 0.028
day 21 , 0.011 day 21 and 0.018 day 21 for 0.1, 1 and
10 mm particles, respectively (Table 3). The gel
permeation chromatograms were unimodal in the
initial rapid decline phase but became multimodal in
the second slow degradation phase, suggesting the
increased production of low molecular weight polymer fractions with time. This trend was observed for
all the particle sizes. The polymer degradation rate
was similar for all the particle sizes during the
second slower degradation phase with the degradation rate constants equal to 0.0080.009 day 21 .
The initial rapid loss in the molecular weight can
be attributed to the dense, compact structure of the
particles, leading to slower outward diffusion of the
degradation products, which can catalyze the degradation of the remaining polymer present in the
particles [12]. This initial loss in the polymer molecular weight has been found to be more rapid for
PLGA polymers with high glycolic acid content.
This is consistent with the more hydrophilic nature
of the PLGA compared to lactide homo-polymers
[13]. The second, slower degradation could be
attributed to the particles becoming more porous,
leading to easier diffusion of the degradation products [12].

3.3. SEM characterization of nano- and

microparticles
The change in surface morphology of nano- and
microparticles with time following incubation in PBS
is shown in Figs. 24. Nano- and microparticles
maintained their integrity for up to 2030 days.
Aggregation of particles followed by matrix-like
structure formation occurred from 30 to 40 days. The
particle surface showed roughness and pore-formation during the terminal phase of the particle degradation. The SEM characterization of nanoparticle
degradation correlated well with the molecular
weight data. The glass transition temperature (T g ) of
the polymer has been shown to decrease with the
decrease in polymer molecular weight resulting in a
more pliable and softer polymer that is prone to
aggregation. This probably resulted in the fusion of
nano- and microparticles seen after 3040 days.
Such plasticization and deformation of the polymer
with hydration has been reported previously. Vishwanathan et al. studied the degradation of microspheres made from different compositions of PLA
and PLGA [13]. Using SEM, they demonstrated that
high molecular weight PLA microspheres tend to
undergo heterogeneous degradation (from surface to
core) and maintain their shape due to their T g being
above 37 8C even after hydration and incubation in
buffer for a prolonged period of time. However, it
was noted that microspheres formulated from low
molecular weight PLAs and PLGAs of any molecular weight or composition tend to undergo deformation and aggregation owing to the lowering of the
polymer T g below 37 8C following its hydration in
the buffer.

3.4. Analysis of surface-associated PVA

Table 3
Values of k (day 21 ) and t 1 / 2 (day) for the degradation of PLGA
particles
Parameter measured

21

k 1 (day )
k 2 (day 21 )
t 1 / 2 (1) (day)
t 1 / 2 (2) (day)
RSQ 1
RSQ 2

Particle size (mm)

0.1

10

0.028
0.008
25
83
0.906
0.995

0.011
0.008
66
86
0.829
0.947

0.018
0.009
39
77
0.918
0.995

Change in the PVA content of nano- and microparticles is shown in Fig. 5. PLGA nano- and
microparticles formulated using PVA as an emulsifier
generally tend to have a certain amount of residual
PVA associated with the particle surface, and this
PVA is unwashable [15]. We have previously shown
that this residual PVA influences a number of
physical and cellular uptake characteristics of PLGA
nanoparticles [15]. In this report, we investigated the
effect of particle size on the amount of residual PVA,

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

179

Fig. 2. SEM of 0.1 mm nanoparticles following in vitro degradation over a period of 90 days: (A) 0 day; (B) 10 days; (C) 30 days; (D) 50
days; (E) 70 days; (F) 90 days.

180

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

Fig. 3. SEM of 1 mm nanoparticles following in vitro degradation over a period of 90 days: (A) 0 day; (B) 10 days; (C) 30 days; (D) 50
days; (E) 70 days; (F) 90 days.

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

181

Fig. 4. SEM of 10 mm nanoparticles following in vitro degradation over a period of 90 days: (A) 0 day; (B) 10 days; (C) 30 days; (D) 50
days; (E) 70 days; (F) 90 days.

182

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

Fig. 5. Effect of size on the amount of (A) residual PVA

associated with degrading nano- and microparticles and (B) PVA
appearing in the buffer at different time points. Each data point is
a mean of three values.

and the influence of residual PVA on the polymer

degradation profiles. The residual PVA content of
nanoparticles (0.1 mm) was found to be generally
higher than the larger size microparticles (Fig. 5A).

This trend was maintained throughout the time

period of the release study. The amount of residual
PVA in the nanoparticles following incubation correlated well with the amount of PVA appearing in the
supernatant (Fig. 5B). The amount of PVA released
into the incubation medium was higher for 0.1 mm
particles compared to the other sizes. The higher
amount of PVA associated with 0.1 mm particle
could influence the degradation of the polymer and
also the release of the encapsulated protein. Surfaceassociated PVA could form a barrier to the penetration of the solvent into the particle leading to
lowered degradation. However, it could also form a
barrier to the release of acidic monomers and
oligomers, which catalyze polymer degradation [10].
Thus, PVA could act to either decrease or increase
the polymer degradation rate. Also, the fractional
PVA released into the medium could influence the
polymer degradation. While there are no reports on
the influence of PVA on PLGA degradation, the PVA
molecule being a large hydrophilic molecule, could
swell in water [19] and thereby reduce the availability of water molecules for PLGA degradation. Since
PLGA degradation proceeds mainly through hydrolytic cleavage of the polymer chains [10], this
reduced availability of water molecules could lead to
reduced polymer degradation. However, in our
studies, the fractional PVA released is insignificant
(in microgram quantities) as it forms a gel at .10%
w / v concentration and hence is not expected to
contribute significantly to polymer degradation.
The influence of residual PVA on polymer degradation could explain the results obtained in our
study, which are in contrast to that of some of the
other previously reported studies on the effect of size
on polymer degradation. Our studies indicate that the
polymer degradation rate of 0.1 mm nanoparticles is
faster than microparticles during the initial degradation phase. On the other hand, Dunne et al.
reported that the polymer degradation rate is slightly
faster for larger-sized microspheres (degradation rate
constant of 0.0568 day 21 for 22.5 mm diameter
particles compared to 0.0447 day 21 for 6.8 mm
particles and 0.0251 day 21 for 0.5 mm particles)
during the initial decline phase [12]. This was
attributed to the fact that in the smaller particles, the
degradation products can diffuse easily to the surface
whereas in the larger size particles, the degradation

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

products have a longer path to the surface. However,

it was found that the particle size did not affect the
polymer degradation rate in the second, slower
degradation rate. This is similar to the results obtained in our studies. Differences in the formulation
procedures and the formulation itself could have
given rise to some of the observed differences. In
their studies, particles were prepared by simple oilin-water emulsion technique without any encapsulated therapeutic agent [12]. In our studies, we used
multiple-emulsion solvent evaporation technique to
formulate all size particles and BSA was used as a
model protein. More importantly, the particles were
formulated with PVA of lower concentration (0.27%)
and lower molecular weight (12,00014,000 Da)
compared to the PVA concentration (see Table 1) and
molecular weight (30,00070,000 Da) used in our
study. Since the amount of residual PVA in PLGA
nanoparticles depends on its concentration during the
emulsion formation as shown in our previous studies
[15], it is possible that there was a significantly
lower amount of residual PVA in microspheres
prepared by Dunne et al. Consequently, the higher
amount of residual PVA associated with our formulation of nanoparticles could form a barrier to the
outward diffusion of the acidic oligomers and monomers, leading to a higher degradation rate. PVA,
present on the surface of particles, has been previously shown to alter the release of encapsulated
therapeutic agents from PLGA nano- and microparticles [6,20]. During the second slower degradation
phase, much of the residual PVA is lost from the
particles, and hence, PVA may not play a significant
role during this phase.

3.5. In vitro release of encapsulated BSA

The release profiles of BSA under in vitro sink
conditions from different size particles are shown in
Fig. 6. The protein release profiles suggested an
initial burst release followed by a lag phase. For 10
mm and 0.1 mm particles, the lag phase was followed
by a slight increase in the protein release. Protein
release was more continuous from the 1 mm particles. In general, the amount of protein released
from the particles was found to be low, with
nanoparticles (0.1 mm) and microparticles (1 mm)
showing higher cumulative percent protein release

183

Fig. 6. Effect of size on the in vitro release of BSA from various

size nano- and microparticles. Each data point is a mean of three
values.

(20% and 32%, respectively) compared to the 10 mm

microparticles (7%) in 90 days.
The in vitro release of the protein did not correlate
with the polymer degradation rate or with the particle
size in the current study. Such a lack of correlation
between protein release and polymer degradation has
been reported previously by others [13]. It has been
shown that the onset of mass loss of the polymeric
device occurs after the onset of the molecular weight
loss and therefore, it has been suggested that drug
release from PLGA matrices could correlate better
with polymer mass loss rather than the molecular
weight loss [13,21]. However, mass loss experiments
could not be conducted with nanoparticles due to less
than 100% recovery of nanoparticles by centrifugation even at high speed. The recovery of nanoparticles could also change with degradation as particles
could clump together with time. Furthermore, the
simultaneous dissociation of surface-associated PVA
and release of BSA from nanoparticles would add
error to the mass loss.
Also, it is possible that factors other than the
polymer degradation rate influence the release of the
protein from PLGA matrices. Some of these factors
are interrelated in a complicated manner. For example, faster polymer degradation should theoretically
enhance the release of the encapsulated therapeutic

184

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

agent. However, faster polymer degradation could

also result in the accumulation of acidic monomers
and oligomers that could lead to protein aggregation
and non-release of the protein. Similarly, smaller size
of nanoparticles presents a large surface area for the
protein to diffuse out more rapidly compared to that
from microparticles. However, smaller-sized particles have been found to contain more surface-associated PVA, which could present a significant barrier to
the outward diffusion of the encapsulated protein.
Thus, some of these above factors could explain the
lack of correlation observed between in vitro protein
release, particle size and polymer degradation.
Initial burst release followed by a period of no or
low protein release from PLGA matrices has been
reported previously and remains a major problem
[22]. In a study by Vishwanathan et al., less than
10% of the encapsulated protein was released from
PLGA microspheres (|120 mm) during the initial
burst release phase with no protein being released
over the next 250-day study period [13]. In their
study, nitrogen estimation of the microspheres suggested that the remaining protein was present inside
the microspheres as water-insoluble form. This nonrelease of proteins from PLGA matrices has been
reported by other investigators and is attributed to
the degradation and / or aggregation of the protein
during the particle preparation and due to the acidic
microclimate generated within the degrading PLGA
matrix [5]. In another study, it was shown that the in
vitro release medium plays a major role in determining the amount of protein released from the PLGA
matrix. It was shown that when PLGA microspheres
are incubated in PBS, there was a slow and incomplete release of protein (lysozyme), primarily due to
the adsorption of the protein to the polymer. However, there was a complete release of the encapsulated protein when the microspheres were incubated
in acetate or glycine buffers [8]. Thus, protein
adsorption to the polymer and protein aggregation in
the acidic microclimate of the degrading particles
could be some of the factors responsible for the
incomplete protein release from nano- and microparticles [23]. In our previous studies, we have
shown greater amount of protein release from the
identical formulation of PLGA nanoparticles. However, in the previous study, the protein release was
carried out in a double diffusion chamber where the

buffer from the receiver chamber was replaced

periodically [15]. The above studies thus strongly
suggest that the in vitro release conditions also affect
the release of protein from nano- and microparticles.

3.6. Stability of protein

Stability of the protein released from nano- and
microparticles was analyzed using SDSPAGE. All
the release samples demonstrated protein aggregation
as evidenced by the presence of bands corresponding
to high molecular weight aggregates in the gels (Fig.
7). Band intensity analysis (Fig. 8) correlated with
the in vitro release profiles. Comparison of the
unaggregated and aggregated protein bands for different size particles revealed that the 0.1 mm
nanoparticles resulted in higher amount of unaggregated protein but also showed higher amount of
aggregation. The 10 mm microparticles showed
lower amount of both unaggregated and aggregated
protein compared to the other size particles. The 1
mm microparticles resulted in the highest fraction of
unaggregated protein in the release medium.
Previous studies have shown that protein instability in PLGA nano- and microparticles could be due to
various formulation conditions used such as due to
the use of sonication or homogenization during the
emulsification step and / or due to the adsorption of
the protein at the oilwater interface leading to the
aggregation of the protein [22]. These factors lead to
the formation of aggregated protein species within
the nano- and microparticles. Another important
mechanism of protein instability in PLGA matrices is
due to the generation of an acidic microclimate
within the polymer matrix due to the formation of
soluble acidic degradation products (oligomers and
monomers) [24]. It was shown previously that neutralizing this acidic microclimate results in the
stabilization of the encapsulated protein [24]. In our
studies, the aggregation of the protein in the release
medium, which could be influenced by the release of
the acidic monomers and oligomers from the particles, was analyzed. The 0.1 mm nanoparticles
resulted in the formation of dense high molecular
weight bands, which could be because of the higher
polymer degradation rate seen in these particles.
Faster polymer degradation would lead to the rapid
accumulation of the degradation products and the

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

185

Fig. 7. SDSPAGE of BSA released at different time points from 0.1 mm (A), 1 mm (B) and 10 mm (C) particles. Lane 1, molecular weight
markers; lane 2, BSA positive control; lane 3, 5 days; lane 4, 10 days; lane 5, 20 days; lane 6, 30 days; lane 7, 45 days; lane 8, 60 days; lane
9, 75 days and lane 10, 90 days.

consequent acidification of the microenvironment

leading to greater protein aggregation. Comparing
the ratio of unaggregated and aggregated protein
bands suggested that 1 mm microparticles had relatively greater cumulative amount of unaggregated
protein released in 90 days.

4. Conclusions
Polymer degradation was found to be biphasic in

PLGA nano- and microparticles, with an initial rapid

degradation during the first 2030 days followed by
a much slower degradation phase. The 0.1 mm
nanoparticles demonstrated relatively faster polymer
degradation during the initial phase compared to the
larger size microparticles. All the particles maintained their structural integrity during this rapid
degradation phase but were followed by pore formation, deformation and fusion of the particles during
the slow degradation phase. All size particles demonstrated aggregation of the released protein, with 0.1

186

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

direct evidence for the role of surface-associated

PVA on the degradation of polymer.

Acknowledgements
Supported by grants from the National Institutes of
Health (HL 57234 to VL and HL 41663 to RJL) and
the Nebraska Research Initiative, Gene Therapy
Program to V.L. Dr. Levys efforts were also supported by the William J. Rashkind Endowment of the
Childrens Hospital of Philadelphia. J.P. and S.K.S.
are supported by pre-doctoral and post-doctoral
fellowships, respectively, from the American Heart
Association. We would like to thank Ms. Elaine
Payne for providing administrative assistance.

References

Fig. 8. Aggregation kinetics of BSA released from nano- and

microparticles. Band intensities of the (A) unaggregated (corresponding to 66 kDa band) and (B) aggregated (corresponding to
high molecular weight bands) protein were calculated and were
normalized to band area.

and 1 mm microparticles demonstrating greater fraction of unaggregated protein released than 10 mm

microparticles. It appears that the surface-associated
PVA rather than the particle size play a dominant
role in controlling the degradation of nano- and
microparticles. A degradation study with particle
formulations having similar diameter but differing in
the amount of associated PVA could provide a more

[1] J. Panyam, V. Labhasetwar, Biodegradable nanoparticles for

drug and gene delivery to cells and tissue, Adv. Drug. Deliv.
Rev. 55 (2002) 329347.
[2] J. Panyam, W.Z. Zhou, S. Prabha, S.K. Sahoo, V. Labhasetwar, Rapid endo-lysosomal escape of poly( D,L-lactide-coglycolide) nanoparticles: implications for drug and gene
delivery, FASEB J. 16 (2002) 12171226.
[3] M. Igartua, R.M. Hernandez, A. Esquisabel, A.R. Gascon,
M.B. Calvo, J.L. Pedraz, Influence of formulation variables
on the in-vitro release of albumin from biodegradable
microparticulate systems, J. Microencapsul. 14 (1997) 349
356.
[4] A.G. Coombes, M.K. Yeh, E.C. Lavelle, S.S. Davis, The
control of protein release from poly( DL-lactide co-glycolide)
microparticles by variation of the external aqueous phase
surfactant in the water-in oil-in water method, J. Control.
Release 52 (1998) 311320.
[5] G. Crotts, T.G. Park, Protein delivery from poly(lactic-coglycolic acid) biodegradable microspheres: release kinetics
and stability issues, J. Microencapsul. 15 (1998) 699713.
[6] Y.Y. Yang, T.S. Chung, N.P. Ng, Morphology, drug distribution, and in vitro release profiles of biodegradable
polymeric microspheres containing protein fabricated by
double-emulsion solvent extraction / evaporation method,
Biomaterials 22 (2001) 231241.
[7] M. Sandor, D. Enscore, P. Weston, E. Mathiowitz, Effect of
protein molecular weight on release from micron-sized
PLGA microspheres, J. Control. Release 76 (2001) 297311.
[8] G. Jiang, B.H. Woo, F. Kang, J. Singh, P.P. DeLuca,
Assessment of protein release kinetics, stability and protein
polymer interaction of lysozyme encapsulated poly( D,L-lactide-co-glycolide) microspheres, J. Control. Release 79
(2002) 137145.

J. Panyam et al. / Journal of Controlled Release 92 (2003) 173187

[9] M. Takenaga, Y. Yamaguchi, A. Kitagawa, Y. Ogawa, Y.
Mizushima, R. Igarashi, A novel sustained-release formulation of insulin with dramatic reduction in initial rapid release,
J. Control. Release 79 (2002) 8191.
[10] J.M. Anderson, M.S. Shive, Biodegradation and biocompatibility of PLA and PLGA microspheres, Adv. Drug Deliv.
Rev. 28 (1997) 524.
[11] M.A. Tracy, K.L. Ward, L. Firouzabadian, Y. Wang, N. Dong,
R. Qian, Y. Zhang, Factors affecting the degradation rate of
poly(lactide-co-glycolide) microspheres in vivo and in vitro,
Biomaterials 20 (1999) 10571062.
[12] M. Dunne, I. Corrigan, Z. Ramtoola, Influence of particle
size and dissolution conditions on the degradation properties
of polylactide-co-glycolide particles, Biomaterials 21 (2000)
16591668.
[13] N.B. Viswanathan, S.S. Patil, J.K. Pandit, A.K. Lele, M.G.
Kulkarni, R.A. Mashelkar, Morphological changes in degrading PLGA and P( DL)LA microspheres: implications for the
design of controlled release systems, J. Microencapsul. 18
(2001) 783800.
[14] M.F. Zambaux, F. Bonneaux, R. Gref, P. Maincent, E.
Dellacherie, M.J. Alonso, P. Labrude, C. Vigneron, Influence
of experimental parameters on the characteristics of poly(lactic acid) nanoparticles prepared by a double emulsion
method, J. Control. Release 50 (1998) 3140.
[15] S.K. Sahoo, J. Panyam, S. Prabha, V. Labhasetwar, Residual
polyvinyl alcohol associated with poly ( D,L-lactide-co-glycolide) nanoparticles affects their physical properties and
cellular uptake, J. Control. Release 82 (2002) 105114.
[16] M.P. Desai, V. Labhasetwar, E. Walter, R.J. Levy, G.L.

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

187

Amidon, The mechanism of uptake of biodegradable microparticles in Caco-2 cells is size dependent, Pharm. Res. 14
(1997) 15681573.
M.P. Desai, V. Labhasetwar, G.L. Amidon, R.J. Levy,
Gastrointestinal uptake of biodegradable microparticles: effect of particle size, Pharm. Res. 13 (1996) 18381845.
S. Prabha, W.Z. Zhou, J. Panyam, V. Labhasetwar, Sizedependency of nanoparticle-mediated gene transfection:
studies with fractionated nanoparticles, Int. J. Pharm. 244
(2002) 105115.
F. Urushizaki, H. Yamaguchi, K. Nakamura, S. Numajiri, K.
Sugibayashi, Y. Morimoto, Swelling and mechanical properties of poly(vinyl alcohol) hydrogels, Int. J. Pharm. 58
(1990) 135142.
M. Shameem, H. Lee, P.P. DeLuca, A short term (accelerated
release) approach to evaluate peptide release from PLGA
depot-formulations, AAPS PharmSci. 1 (1999) E7.
I. Grizzi, H. Garreau, S. Li, M. Vert, Hydrolytic degradation
of devices based on poly( DL-lactic acid) size-dependence,
Biomaterials 16 (1995) 305311.
M. van de Weert, W.E. Hennink, W. Jiskoot, Protein instability in poly(lactic-co-glycolic acid) microparticles, Pharm.
Res. 17 (2000) 11591167.
S.P. Schwendeman, Recent advances in the stabilization of
proteins encapsulated in injectable PLGA delivery systems,
Crit. Rev. Ther. Drug Carrier Syst. 19 (2002) 7398.
G. Zhu, S.R. Mallery, S.P. Schwendeman, Stabilization of
proteins encapsulated in injectable poly (lactide-co-glycolide), Nat. Biotech. 18 (2000) 5257.